© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 20342
    Prev
    Next
    INDEX
    Explanations

    catchment, reflector, ItemGroup

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron activates on domain-specific technical nouns—most strongly on words like “catchment” (and “catchments”) or “reflector” (and “reflectors”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    鱚
    -2.59
    It
    -2.45
    　　-
    -2.39
    鶲
    -2.34
    萄
    -2.31
     lyd
    -2.31
    뜸
    -2.30
    
    -2.27
    了
    -2.27
    ！”
    -2.25
    POSITIVE LOGITS
     '
    2.59
     otras
    2.53
    ',
    2.50
    л
    2.42
     
    2.42
     insanlar
    2.36
     --
    2.34
     самая
    2.33
     najbardziej
    2.31
     wonderful
    2.28
    Activations Density 0.005%

    No Known Activations